[Discussion] Development direction

[Ekdohibs]

Minetest Game is as of now the only game that is bundled with Minetest. Thus, it should be made attractive to newcomers (actually, I would say that Minetest Game needs to be developped more, and that making a basic game is for minimal). Now, looking at what most newcomers want:

• [ ] Mobs. This is the highest one on the list, but also the most problematic one. So we should decide what we want to add - but it should be lightweight (so no abm over air or things like that); while for them to be fun enough, there should be some variety (maybe at least 2 peaceful mobs and 2 hostile @ones).
• [ ] Mapgen
  • [ ] More underground variety - #1944
  • [x] More biomes - Right now, mapgen, is a bit boring (only 4 biomes...), so more biomes and biome-specific resources would make the game more interesting; it would also make the player travel a bit more.
• [x] Ambient/environmental sounds - to make the world seem more alive. Needs engine improvements.
• [x] Weather - Needs engine improvements
• [ ] Craft guide - #1435
• [x] Faster way to travel - Well, if the player has to travel, some way that is quick would be really useful - either teleportation devices or mods such as carts (I prefer the latter because it requires some infrastructure, and because the longer the distance, the more expensive it is).
• [ ] Objectives - Right now, Minetest Game has no objectives - even surviving is easy since there is almost nothing that can kill you (just lava and falls). Objectives would be a key point to make gameplay more interesting.
• [ ] Automation / Circuits - I'm not saying we should include mesecons or pipeworks - but perhaps some parts of at least mesecons (just the basic things - the pressure plate + door combination is used by many players)

[kilbith]

If you add such high variety of biomes, be sure that they're large enough so that the world is not too "mosaic" in biomes.

[nerzhul]

@Novatux i agree with all of your points.

[C1ffisme]

You know, I was playing minetest earlier, and I've noticed that one of the first things I do is go right down and start mining. I can get iron fast if I find some near the surface. But underground is rather boring. All the caves just look the same. Covered in the same ores, same stone, same stuff.

I thought of an idea, what if mining had a sense of exploration? We could add dungeons with chests and mobs, ravines as a change from circular caverns, and even underground fortresses, with ancient technology that you can not only re-use, but copy and re-construct! Imagine searching for a stronghold underground, getting ready to find a powerful weapon to complete your house's defense system.

Also, we could make underground biomes like Terraria! Forget just seing stone, and stone, and more stone, we could add a nether biome, a jungle biome, and all different kinds of stony biomes! You could have forests surviving off of lava light, a nether with pools of lava, granite caves with pools of mercury or some chemical or even molten steel! Just imagine taking the interesting mapgen above into the resourceful world below! Remember, we don't have to copy reality!

And what about a fight? Or maybe even underground battles? You could build metal robots with the defense systems you find underground, and attack zombies or other strange lifeforms found under the earth. You could fight aliens that dug into the earth to steal your minerals, kill protective guards that won't share their mese, and burn down anyone who decides that you don't deserve their materials!

Those are just some ideas. But lets get back to the point. Underground is boring. We are called Minetest, are we not? So, let's make mining in Minetest be a little less about seeing who can gather the most mese, and more about who can steal the golden laser turret from the dungeon without getting killed by the stone-bots/zombies/whatever crazy mob you can think of!

[paramat]

If you add such high variety of biomes, be sure that they're large enough so that the world is not too "mosaic" in biomes.

Yes because of the higher number of biomes i increased heat and humidity spreads to 1000 nodes.

[thatgraemeguy]

@C1ffisme: the answer, as always, is that all that stuff can be done in mods. minetest_game is staying relatively simple, I would've thought that much was clear by now.

[C1ffisme]

@thatgraemeguy Alright then. Let's just stop adding features right now. We'll let every feature be taken care of by mods.

1 year later...

Minetest is completely forgotten. Everybody didn't bother with it because nothing official came out, you just modded it.

If we just leave things to mods, then nobody likes minetest_game. If not changing minetest game, at least we should make a new one so minetest dosen't become boring. Everyone loves news, and I think it would be nice if every so often we heard of updates and new blocks, etc.

[rubenwardy]

We don't operate like that. If we added new blocks every release, the game would become incredibly bloated, and you can't take back items after you add them. Minetest is meant to be modded, so the current philosophy is to make the default only a foundation in order to encourage this.

This sort of conversation is much better off on the forums.

[thatgraemeguy]

:+1:

[asl97]

i like @C1ffisme idea but minetest_game is suppose to just be the base.

there should be an official (bloated) playing game and this should be the official modding base.

@C1ffisme if you create such a game, i would try it and i am sure there would be plenty of people who would want it too.

[C1ffisme]

I might have something soon, I'm working on a game slightly different to most voxel games, though I might need help with graphics.

[C1ffisme]

If this is the direction minetest_game is going, then we need another game quick. Right now it looks like minetest_game is on a failure mission for people who aren't programmers, don't mind the bad gameplay balance of mods, or can somehow survive without goals.

How many of you people actually play real games designed for fun?

[paramat]

^ That's unhelpful.

Ticked the mapgen box.

[0-afflatus]

@paramat

Ticked the mapgen box.

You have indeed. Thank you. I'm particularly enjoying the biome boundaries.

[Ferk]

Yes, my regards as well, mapgen v7 is pretty nice.

About the other checkboxes.. I don't think there needs to be a gameplay goal if the (unclear) project goal is to be a sandbox. Games/mods could add more interesting different gameplay goals.

The main problem with the other 3 points (mobs, railcarts and pressure-plates) is the increased load for the server and the fact that latency plays a big role in these features so they are bound to be laggy. A client-side API might help... not sure if there's much activity going on in that direction.

[C1ffisme]

Sorry, I was just... aggravated. I'm tired of seeing minetest not having anything to do or see. The blocks are almost all the same as minecraft. Grass, stone, dirt, there isn't anything new or better than minecraft except mese and nyan cats. I want to see minetest become an actual fun game to play, not just a tool for creative art or PVP. The vanilla game just isn't... fun... enough. With my new laptop, I'm making a new minetest_fun game with cool power tools and my own mobs and bow/arrow mods.

Why is it we just won't add things? I see tons of pull requests, but almost all are bugfixes or minor changes. And anything new isn't a new metal or tool, it's a biome. I see no reason why we can't implement things such as Titanium or Power drills? I would even welcome just a few more crops.

What I think would make survival feel more fun is if we added hunger. It would mean at least one reason to do something in the game. Then at least people are required to build farms and storehouses. Add cool machines to refine food, and people will build cool factories.

[Ferk]

Hunger (and/or aggressive mobs) would kind of add sort of a goal: to survive

Specially if the rate satiation decreases is customizable.. you could either make a very extreme survival requiring food constantly or opt-out of the mechanic by making you always full at all times.

Though.. opting out this way would also make you permanently gain a regeneration boost, unless either we figured out a different way to gain health or made a fallback to the current behavior of food healing directly. But then... why would you add a lot of optional behavior instead of just cleanly using a separate mod? ..you can toggle mods easily if they are not included in minetest_game, but getting rid of a mod that is included is not as simple.

I assume this is the reason it wasn't added.

Personally, I would have just added hunger anyway (I find this way more important than other optional features like TNT, which I would have personally not included) along with a setting for the rate the satiation decreases and just have different items for healing than food (maybe have regeneration but at a much slower rate).

[Dragonop]

I think I said this some years ago, but what if we bring the 2d mobs back? Those were really fun for me at least, and we already have all of them (2 hostile, dungeon master and oerkki, and 2 friendly, firefly and rat). What was the problem with them and why did that got removed, could you tell me (again) please?

[C1ffisme]

@Dragonop I think the problem with bringing the 2D mobs back is that they would still have to be written in lua, since hardcoding them into the engine isn't exactly a good idea. (I myself am making a 'realistic' minetest subgame, and oerkkis and dungeon masters aren't very realistic.)

A mob API, with built-in Artificial Intelligences, commands, etc. is the way to go. Current mob mods will have hacky implementations inside of their AIs, causing the server to lag as node scans or large calculations multiply with every mob. However, if the calculations/scans were added into the C++ code, making it much faster, we could have a chance at mobs in Minetest Game.

[rautars]

Disclaimer: this is not a rant, only perspective from guy who recenlty joined minetest community.

For now lack of biome is least problem. Unclear minetest focus - thats a critical problem.

Default mod is already bloated 6 - same type tools, 6 - same type swords 0 - armor. Why swords even included when there no enemies (pvp-only eddition)?. Tons of blocks which differs only color. Some WTF list: nyan cats, obsidium glass, locked/unlocked chests (why two same objects). Item rails defined in 'default' but to actualy use rails, carts mod needs to be installed.

From my point of view minetest target audience is two groups:

• people who wants download and play
• people who set-up server and heavily tweaks it by mods

Judging from minetest game content it unclear focus. Some items make sense for server (pvp? swords, locked chests, bones) and some items which does not (water/lava bucket). Frustration comes even in creativity game when playing as singleplayer still need grant yourself flying privileges. When trying build something its pain to search items. Some items even not visible in dark inventory slot backgroun (poor UX design), item order is unclear, hard to find items.

If most popular minetest subgame can't implement things rightly, how you expect solo developers make better subgame by themselfs? They will burn-out and will leave what they started (already lots of half baked mods).

Userbase will not grow with 'minetest game' and with poor userbase it will be lack of mod creators. It obious how it will end up.

If minetest devs believe focus only on api model is right way then probably after almost year there should be plenty subgames ready to include into 'new world' game selection list for Minetest newcomers.

[raymoo]

Is it ok to make a separate issue for features discussed in this issue?

[PilzAdam]

@raymoo yes, this issue is a mess and nobody reads it anyway.

[0-afflatus]

@xeranas I think you will understand some of this better when you have played the game a bit more. There are 89 sub-games at last count and almost too many willing mod creators. We could probably do with more online players than anything else. Part of the beauty of this game is its open-endedness and I think it is appropriately reflected in the development. In my experience the best games are those that have simple rules with many possible outcomes.

[rautars]

I think you will understand some of this better when you have played the game a bit more

Thats not how games should work. First game impression is very important and if game does not hook up you then its high change you won't play it anymore.

@0-afflatus walking around with MT fanboy glasses does not helps project in any way. Issues needs to be addressed and fixed.

[RHRhino]

@xeranas You are right and wrong: Yes the first impression of a game is important and if people don't like the game they won't play it, but also keep in mind that minetest is a open source game engine in development with a playable (also in development) game and not a finished product like a polished triple-A video game.

[ElementW]

minetest is a open source game engine in development with a playable (also in development) game and not a finished product like a polished triple-A video game.

@RHRhino But if you're not heading towards that AAA-like feeling, or any fluidity in UX whatsoever, then you're going nowhere.

I could rant for literal hours about what's wrong with MT, and this very lack of any ambition is what I'd reproach all devs of.

[0-afflatus]

@xeranas all games have a learning curve. You appear to have made quite a lot of assumptions based on ignorance and a rather fixed attitude about how games should be. I am aware of MT_game's failings, that's why I'm building a sub-game (which contradicts one of your arguments). Some of what you say is useful, I just think you need to familiarise yourself with the territory. Then you will understand why there are two chests, one locked, one unlocked and then perhaps you will be able to submit a workable suggestion of how to change that.

This conversation has been running for some time on various different threads and many of the issues are being addressed, hence we now have an excellent biome system.

My answers to other points of the OP

Mobs: mobs_redo without all the add-ons. I actually don't see the problem with this, but I do understand it is not to everyone's taste.

Automation: I would like to see and when I get time develop a simple machines mod, based on Archimedean principles ... and something like a pressure plate.

Travel: horses and carts, yes. I also use /tpr in my game, because it works.

The only thing I haven't got an answer for at all is goals. I enjoy the open-endedness and I wouldn't want to go down the path that MC took. I understand that this is important for many players, so I am working on a quest system.

Talking about ambition is a bit meaningless to me. I think your complaint @ElementW is that they don't share your ambition. We all want to make the game as good as it can be, but that takes time and patience. Sniping on the sidelines is easy.

[paramat]

We're planning to add a version of smalljoker's 'boost carts' mod, see #1311

[paramat]

Thanks to sofar i have now ticked a second box: 'Some way to travel quickly'.

[C1ffisme]

@paramat I won't say that one method of transport will solve all player's needs, but it works for now.

I feel as if when I play Minetest, I feel as if little things are bugging me, such as the lack of ambience, or just the fact that some textures are too smooth. Many Minecraft textures are much more contrasted, and have a bit of a retro feel to them. This might have also been what kept me entertained so much with Minetest back in 2012.

A screenshot of 2012 Minetest: (Sure, those textures may look ugly to some, but they still looked like what they represent as blocks, and I think they did their job well.)

A screenshot of 2016 Minetest: (These textures lack the same contrast, and almost feel 'dustier' in a way.)

I think a combination of both retro, cute graphics combined with some smoother, more modern graphics would be more appealing. Perhaps some of the newer textures could have some brighter colours?

[RHRhino]

@C1ffisme Minetest and Minecraft have two different styles of pixel art. Minetest is aiming a smooth natural looking style, while Minecraft aims retro looking pixel art.

I think a combination of both retro, cute graphics combined with some smoother, more modern graphics would be more appealing.

There's a rule in art, that you should stick to one style instead of mixing different styles, because most people sense it more harmonic if you stick to one style. That applies to the art style (smooth natural, retro, minimal) as well as color and texture size - they all should stay consistent. Unfortunately Minetest already has some small inconsistence, because the textures were changed a lot.

In general I wonder why so many people complain about textures, because they are probably the easiest thing to mod for yourself if you don't like them. It should be obvious that minetest_game can't provide textures that everybody loves - some people will always prefer other textures, although I agree some texures could do with some further improvements.

[C1ffisme]

@RHRhino Maybe I don't know what I'm talking about. I dunno. It's hard for me to accurately pinpoint what I find wrong with Minetest's default textures. But maybe certain graphical effects would help as well, since even Minetest textures in Minecraft look okay:

This one is going to be a bit less pleasant, because of the Andesite and Granite that contrast with Minetest stone more than Minecraft stone, but it still looks fairly good:

[paramat]

The default textures are just the default, they change very slowly one by one and with many different authors, and will continue to, so some lack of consistency is inevitable. Considering this they are surprisingly reasonable. It's inevitable and expected that most players will not like them and will prefer to use texture packs instead.

[C1ffisme]

As another comparison, I decided to recreate a Minecraft screenshot in Minetest. It's not exactly the same, if you look through the window, but here you go:

Minecraft:

Minetest:

Two things pop out at me the most: (Ignoring the other HUD elements for XP, hunger and health, and the background seen through the window.)

• The lighting
• There is a bit of stretching at the sides of the Minetest Screenshot.

[sofar]

stretching

seems both FOV and aspect ratio are different, so the screenshots are not comparable in that sense.

[paramat]

C1ffisme these are non-issues, the games look different obviously, and just use a texture pack you like.

[C1ffisme]

@paramat I want to find what makes these games' atmospheres so different. No matter what texture pack or font I use, the game still looks too much like a graphical test. :cry:

[raymoo]

@HybridDog That happens in Minecraft too.

[paramat]

MC will inevitably seem more polished, due to their human resources.

[rubenwardy]

updated OP

[paramat]

I strongly disagree with these statements:

Right now, Minetest Game isn't a real game because you lack goals [...] So a goal - there are plenty of possibilities, but we have to chose one

Thinking a game needs a goal is mainstream conditioning. I followed the development of MC from beta days, it was a wonderful game being open-world and open-ended, but then they felt forced to add 'the end' which is out of character with the rest of it. Anything that makes a player feel like they have 'finished' is a bad thing in MT. We can have objectives but there doesn't have to be just one.

Concerning mapgen stuff, the tickboxes would be better as:

• Add underground decorations: plants, crystals.
• Underground biomes to vary those decorations.

Please can underground rivers be removed? These are only possible through engine mapgen code and cannot be created by a subgame. We already have water caves, where these intersect a tunnel water will flow down the tunnel.

[paramat]

Edited first post.

[C1ffisme]

@ElementW I'd be interested in seeing a gist with a rant in it, if only for the motivational value.

@paramat

Please can underground rivers be removed? These are only possible through engine mapgen code and cannot be created by a subgame.

Unless I'm dead wrong, can't you do any form of map generation using minetest.register_on_generated()?

[Fixer-007]

I will provoke everybody:

Minetest Ultimate Goal = Minecraft minus Enchanting/Potions/Elytra/... minus Realms (End, Nether, etc) plus Our Own Stuff (tm)

[rubenwardy]

This post is about the direction of the game play, and so it's okay imo to have engine issues here / needs engine change

[C1ffisme]

@Fixer-007

Minetest Ultimate Goal = Minecraft minus Enchanting/Potions/Elytra/... minus Realms (End, Nether, etc).

But with the mod API able to add these features if wanted. (Although, if we added the ability to make realms, we might as well create one for MTG anyway, either as an example or to enhance gameplay.)

[paramat]

and so it's okay imo to have engine issues here

Yes it is, sorry, i removed underground rivers more because it was a random idea and we should focus on a smaller number of widely agreed intentions.

Unless I'm dead wrong, can't you do any form of map generation using minetest.register_on_generated()?

Yes but such Lua mapgen is too slow to be used for default MT mapgen, much better to code it in C++.

[C1ffisme]

Yes but such Lua mapgen is too slow to be used for default MT mapgen, much better to code it in C++.

Well, of course it would be too slow for default. I was just pointing out that they weren't impossible to do.

[numberZero]

Minetest is a nice engine, but minetest_game is not a game, it’s a modding framework. It should be kept as such for experienced players.

But it’s probably a good idea to just bundle a set of mods with Minetest. These should not be considered an integral part of the game, but should be immediately available (and probably enabled by default) for newcomers. I would include some mobs (don’t know any mod good enough actually), mapgen enhancement (like ethereal or even mg) to make it look better, technical stuff like moreblocks, probably moreores and moretrees and, surely, something like MinetestAmbience, if there would be a not-so-laggy version. Sounds are really important for a game.

We actually have almost all we need in mods, just visit some good server like Survival in Ethereal to see it, it looks very nice. But such nice look should be available in freshly downloaded game.

the game still looks too much like a graphical test.

Agree partially. But some subtle things, like tone mapping, make it look A LOT better.

[C1ffisme]

I thought I'd throw this out there, and see what everybody else thinks.

Currently, the game's progression is... incredibly bland, not to mention very linear. The most exciting thing about the progression is that you can get a bronze pick rather than a steel one:

Minecraft's progression is longer, sure, but it's more than just length that keeps it interesting. You can skip getting a diamond pickaxe to get to the nether by using two buckets to "mold" a portal. Then after that, it's not just about getting a new tier of pickaxe. You explore the nether to find a nether fortress, which you then fight blazes to get blaze powder, which allows you to get to the end. From the end you can fight the ender dragon, allowing you to get to the end islands where the elytra, which is the most endgame item I can think of, can be found.

And this is only including the main progression line. There are all kinds of other things you can do. Once you have an iron pick, you can mine and use redstone, if you find a nether fortress, you can brew potions, or fight the wither so you can make a beacon. You can also skip lots of early-game progression using villager trading, which is it's own aspect of the game.

My point? Progression is more than just the next tier of tool material, it should also fork off into things you can do with those materials (Other than just for decorative purposes.) and should be more creative than just "Find something underground that you can make a tool out of", including surface exploration and even requiring creative problem solving (Something that MC's main progression lacks, but does help with say, automation).

EDIT: Hee hee, I forgot an arrow between the diamond pickaxe and nether portal. Oh well.

[Iniquitatis]

@tobyplowy Completely agreed.